Browse categories
Explore
Fiverr Pro
English
$
USD
I will build ai chatbots and rag development
Awais
About this gig
Want an AI chatbot that actually knows your business not a generic GPT wrapper?
I build production-grade Retrieval-Augmented Generation (RAG) chatbots that answer questions using YOUR documents, website, or database with accurate, source-grounded responses.
I recently built this exact kind of system for a bank client: a chatbot that crawled 80+ pages of internal content, embedded it into a vector database, and answered employee questions in under 10 seconds running on a self-hosted LLM for full data privacy.
What I can build for you:
- Chatbots grounded in your PDFs, website, Notion, or database (no hallucinations)
- Integration with OpenAI, Claude, Llama, or other open-source LLMs
- Vector database setup (ChromaDB, Pinecone, FAISS)
- REST API backend (FastAPI) + simple chat UI (Streamlit/web widget)
- Self-hosted/local LLM setup if you need full data privacy (no data leaves your servers)
Tech stack: Python, LangChain, Hugging Face, ChromaDB, FastAPI, OpenAI/Llama, Streamlit
Message me with your use case before ordering if you're unsure which package fits happy to advise.
Get to know Awais
Awais
AI Engineer LLM, RAG Chatbots Computer Vision
- FromPakistan
- Member sinceMay 2021
- Avg. response time1 hour
Languages
English, Urdu
AI Engineer with 1+ years production experience building LLM chatbots, RAG pipelines, and computer vision systems for clients in retail, FMCG, and banking. I turn messy data into deployed products — YOLO detection pipelines, bank-grade RAG chatbots with sub-10s responses.
• End-to-end: preprocessing → model → API → UI
• Real impact: cut manual task time 60%, saved teams 4+ hrs/week
• Stack: LangChain, Hugging Face, FastAPI, ChromaDB, OpenCV, TensorFlow
• Fast, clear async, cross-time-zone communication
Let's discuss your project — I'll tell you honestly if AI fits first.
FAQ
Do I need to provide an OpenAI API key?
Yes for OpenAI-based bots, or I can set you up with a free open-source LLM (Llama) if you'd prefer no ongoing API costs.
What if my data is sensitive/confidential?
I offer fully self-hosted setups (local LLM + local vector DB) so no data ever leaves your infrastructure.
Can this integrate with my website?
Yes — I can deliver an embeddable widget or a REST API your dev team can plug into your site.
Do you sign NDAs?
Yes, happy to sign an NDA before starting.
